Hedge Funds’ Returns: An Alternative Investment or An Alternative to Investment? (Part 2: Dealing with Biases)
by Dr Constantin Gurdgiev, Adjunct Assistant Professor of Finance with Trinity College, Dublin
You may find Part 1: Measuring What here, Part 3: Dealing with Funds and Benchmarks Selection here and Part 4: To higher moments and beyond here..
As I noted in the previous post1, the problems with the top-level view of the hedge funds driving both the negative assessments of their performance and the volatile nature of in- and out-flows of cash they experience are multiple. The core CFA Alternative Investment and Investment Theory modules across all levels of CFA body of knowledge attempts to address some of the main methodologies used in assessing the risk-adjusted returns for the hedge funds. However, even equipped with the latest data analysis models, an analyst of the funds performance should be aware of several core biases that often impact the sector.
Risks v Returns: A Measurement Bias
If the hedge funds are indeed pursuing risk-adjusted returns (at least in those funds that have a linear alignment between own returns and clients’ returns), then any discussion of the hedge funds relative performance, compared to other asset classes should cover not only actual returns, but also returns volatility and higher moments of returns distribution, most notably skewness, co-skewness, kurtosis and co-kurtosis.2 One recent study by Kolokolova and Mattes looked at the determinants of risk for highly liquid hedge funds.3 The authors found that “larger funds and funds charging higher incentive fees exhibit lower risk, whereas funds charging higher management fees, imposing longer notice periods, and stemming from large fund-families take more risk”. In part, this confirms findings concerning the negative correlation between hedge funds management charges and funds performance (see the discussion referencing Lajbcygier and Rich, 2014 paper below.4
But Kolokolova and Mattes also found that hedge funds performance is not uniform across geographies and currencies in which the funds operate. And they are also heterogeneous across core macroeconomic risks. Nonetheless, as an asset class, hedge funds do generate significant betas for core shocks, such as default (positive beta) and inflation (negative beta).5
Another illustration of the importance of risk-adjustment in consideration of funds’ returns is provided by Maziar Kazemi and Ergys Islamaj (2014)6. The authors looked at the returns on active hedge funds investments compared to more passive strategies. Controlling, partially, for the survivorship bias, the authors found that “there exists a nonlinear relationship between activeness and performance. Using raw returns as a measure of performance, …more active funds outperform the less active ones. However, when risk adjusted returns are used to measure performance, we find the opposite results; that is, activeness is inversely related to returns.
Timing Risk and Strategy Variations
As with all assessments of returns to broader asset classes, one has to be careful not to fall into the trap of choosing the wrong time horizon for consideration. Let me explain briefly what this means. At any point in time, some asset classes outperform others and some under perform all or majority of investments. Currently running example on the upside: U.S. and European equities and U.S. and European bonds – all of which have been posting historical highs in recent months despite widely divergent underlying fundamentals. On the downside: commodities, especially oil. Media coverage tends to focus on two tails of the returns distribution within a snapshot of time, namely today. The right tail, where the assets returns are the highest and the left tail, where the assets returns are the lowest. But this relative performance positioning reverses over time and often the mid-range of returns asset classes move either left or right in the tail. Instability of even the medium-term predictable relationships (such as high performance or low volatility of specific asset returns) is a feature of the markets, even though it is rarely a feature of the media coverage.
Correcting for the benchmark biases whether induced by the choice of time horizon or strategy, an alternative view of the hedge funds returns can be found in assessing funds’ absolute returns.
A number of authors attempt to do so using various econometric techniques. For example, Tudor and Cao (2012)7 found that funds do generate absolute returns, but the returns consistency and even the likelihood of generating them are different across various strategies. The authors found “no evidence … for performance persistence in terms of absolute returns for hedge funds but some limited evidence for funds of funds.”
In summary, hedge funds’ returns should not be measured on a relative basis absent risk considerations and without considering the returns variation relating to time horizons chosen to cover these returns. As hedge fund investment exposures are usually held by institutional investors, pure returns on funds can obscure both the risk management properties of these assets and the timing / liquidity constraints faced by institutional holders.
Resources: Social Science Research Network, IDEAS , CFA Institute and The Journal of Financial Economics.
References
2. See, for example, CFA publication on the matter
3. Kolokolova, Olga and Mattes, Achim, “How Risky are Low-Risk Hedge Funds?”, October 10, 2014:
4. Lajbcygier, Paul and Rich, Joe, “Do Incentive Fees Signal Skill? Evidence from the hedge fund industry”, July 1, 2014:
5. Bali, Turan G., Brown, Stephen J. and Caglayan, Mustafa Onur, “Do hedge funds’ exposures to risk factors predict their future returns?” Journal of Financial Economics, Volume 101, Issue 1, July 2011: pages 36-68
6. Kazemi, Maziar and Islamaj, Ergys, “Returns to Active Management: The Case of Hedge Funds”, August 19, 2014. FRB International Finance Discussion Paper No. 1112
7. Tudor, Deniz and Cao, Bolong, “The Absolute Returns of Hedge Funds”, Managerial Finance, 2012, Volume 38, Issue 3: 280-302